NVIDIA releases open-source transcription model Nemotron 3.5 Transcribe
What happened
On 28 October 2026, NVIDIA released the open-weight speech model Nemotron 3.5 Transcribe, available on Hugging Face and ModelScope. According to the Hugging Face blog, the model combines multilingual transcription, speaker diarisation, translation and audio question answering in a single checkpoint, and NVIDIA also released the Granary v2 data pipeline. TechCrunch reported the same day that the model can transcribe audio in up to 40 languages, translate between 31 languages and English, handle recordings up to one hour long and distinguish up to eight speakers; NVIDIA says it outperforms competitors on benchmarks including the Hugging Face Open ASR Leaderboard. NVIDIA's blog also announced several updates for RTX Spark laptops the same day, including the model, support for Liquid AI's LFM3 series models, and a control in the NVIDIA App for allocating GPU resources between foreground applications and background agents. The Decoder later added that the model can be controlled with text instructions and supports speaker identification in 13 languages. The model has now been released; the reports do not mention third-party validation.
Written by AI from the reports · updated 34 min ago
Developments
- Oct 28 23:00 · 1 reportsNVIDIA releases Nemotron 3.5 Transcribe speech model and RTX Spark updatesNVIDIA Blog：NVIDIA RTX Spark Windows Laptops Gain New Models, Agent Tools and Games
- Oct 28 22:58 · 3 reportsNvidia releases open-source transcription model Nemotron 3.5 TranscribeHugging Face Blog：NVIDIA Nemotron 3.5 Transcribe and Granary v2: Intelligent Multilingual Transcription You Can Prompt and Adapt
Timeline
Follow the reports to see the event from each side.
- The DecoderNvidia's open Nemotron 3.5 Transcribe combines transcription, translation, and speaker recognition
Nvidia has released Nemotron 3.5 Transcribe, an open speech recognition model that handles transcription, speaker recognition, and translation and takes its instructions as text prompts. It leads several benchmarks and uses only a sixth of the active parameters of its strongest open rival.
- NVIDIA BlogNVIDIA RTX Spark Windows Laptops Gain New Models, Agent Tools and Games
NVIDIA RTX Spark powers thin and light laptops built for everything that developers, creators and gamers do. RTX Spark laptops are now available from ASUS, Dell, HP, Lenovo, Microsoft and MSI.
- TechCrunch · AIVoice AI in vogue: Nvidia releases new open transcription model
Nvidia just released a free speech-to-text model that can transcribe up to 40 languages, but it only works on recordings, not live audio, so it’s a big step for meeting and medical transcription tools while still leaving real-time use to rivals like ElevenLabs.
- Hugging Face BlogSelectedNVIDIA Nemotron 3.5 Transcribe and Granary v2: Intelligent Multilingual Transcription You Can Prompt and Adapt
One open model for multilingual transcription, speaker attribution, translation, and audio Q&A, with top leaderboard results, plus the open Granary v2 data pipeline and fine-tuning recipes to adapt it.
Heat of this event
Current heat 36·Peak in the comparable range 36（Oct 29 03:00）·Change over 24 hours in the comparable range –
The trend compares only the same accounts observed without a break, so its range can be narrower than the current heat count. Move the pointer or tap the chart to see each hour; with a keyboard, use the left and right arrow keys.